Nucleic acid sequence analysis has become a cornerstone in many activities in biology, biotechnology and medicine. The ability to determine nucleic acid sequences has become increasingly important as efforts have commenced to determine the sequences of the large genomes of humans and other higher organisms and also, for example, in single nucleotide polymorphisms (SNPs) detection and screening and gene expression monitoring. The genetic information provided by nucleic acid sequencing has many applications, for example, in drug target discovery and validation, disease diagnosis, risk scoring, and organism identification and characterization.
Due to the rapidly increasing demand of reliable data on the genetic information related to an organism, a disease, or to individuals into a population, it is more and more important to improve consequently the throughput of the sequencing methods. The basic objective of such applications is the determination of the sequence of the four bases adenine (A), cytosine (C), guanine (G) and thymine (T) or uracil (U) comprised in the nucleic acids of interest, and belonging to a cell type, to an organism, or to a population of individuals. Most of the nucleic acid sequencing methods (reviewed in Yan H et al., Science 2000, 289 (5486):1890-2) allows the determination of a nucleic acid sequence, either directly (using primer-driven nucleotide extension or hybridization technologies) or indirectly (by electrophoretic and/or cleavage analysis). However, the scientific and economic value of any application dedicated to nucleic acid sequence analysis is highly dependent from the actual throughput of the method.
The efforts necessary to obtain an isolated nucleic acids in a quantity and of a quality acceptable to have reliable results constitute one of the main issues to be considered when evaluating the throughput of a sequencing method. The original sample containing the nucleic acid to be sequenced often does not provide enough material to perform such analysis, unless either relatively high quantity of starting material and/or considerable human intervention are provided. This is particularly important in the case of forensic or archival DNA specimens, or mRNA samples obtained from particular cell types, when sequencing methods needing hundreds of nanograms of an isolated DNA sequence have to be performed starting from nanograms, or less, of total genomic DNA.
To overcome such problem, the solution mainly adopted is to amplify a DNA sequence by generating several copies only of the nucleic acid fragment of interest. In general, the amplification of a nucleic acid needs a series of actions to be repeated:                a) Providing a first nucleic acid (the template) containing the sequence of interest;        b) Providing a second nucleic acid (the primer) containing, at least at its 3′ end, a sequence complementary to a sequence contained in the first nucleic acid and adjacent, or internal, to the 3′ end of the sequence of interest;        c) Providing the conditions allowing the transition of both template and primer into single stranded molecules, at least in the complementary region of the two nucleic acids;        d) Providing the conditions allowing the hybridization between the complementary sequence located in the template and in the primer;        e) Providing a nucleic acid polymerase able to synthetize the complementary strand of the template region starting from the 3′ end of the hybridized primer following the Watson and Crick base pairing rules (A-T, or A-U, and C-G);        f) Provoking the separation of the resulting two stranded nucleic add (one strand belonging to the original template, the other containing the primer fused to the sequence complementary to the region of the template where the nucleotide extension was possible), so that each strand can become the template for another hybridization and polymerization.        
Usually, a second primer is also added into the amplification reaction to start the polymerization by hybridizing it to a portion of newly synthetized strand adjacent, or internal, as well to the sequence of interest. If two primers, each specific for one complementary strand, are used, the repetition of these cycles allows the synthesis of multiple copies of both strands of the template sequence comprised between the sequences used to hybridize the primers. Alternatively, if only one primer is provided, multiple copies of a single strand are produced.
The key point is to restore continuously the conditions permitting the dehybridization of the original and the newly synthesized strands and their hybridization to the free primer molecules to restart the polymerization and generate quickly the desired amount of the specific nucleic acid sequence. Primers molecules are generally present at a concentration considerably higher than the template molecules, so that the exponential kinetic of amplification process can be supported.
Commonly, the transition between single stranded and double stranded form of the nucleic acids is obtained by raising and lowering the temperature of the system in which the amplification process is performed. At this scope, the elements necessary for the amplification (template, primers, nucleic acid polymerase, nucleotides, salts) are submitted to a series of heating and cooling phases generated by an external device.
Such approach, usually called thermocycling, is applied in the well-known technique called Polymerase Chain Reaction (PCR; Taylor G R, “Polymerase chain reaction: basic principles and automation” in “PCR: a Practical approach”, edited by Mcpherson M J et al., Oxford Univ. Press 1991), in which the following step are conducted in a precise order and for predefined time periods:                a) Denaturation (90° C.-95° C. for 30-60 seconds). The external device provides the energy (in the form of heat) necessary to increase the movement of the nucleic acid molecules at a level sufficient to break the non-covalent interactions stabilizing double stranded configuration. This step can be eventually omitted in the first cycle, if the template and primers are already in a single stranded form, but it is absolutely necessary in all subsequent cycles.        b) Hybridization (35° C.-80° C. for 30-90 seconds). Complementary sequence contained in the primers and in the template molecules can hybridize once that the temperature is lowered. This temperature, which has to be attentively controlled to avoid the hybridization of molecule having only limited homology, is calculated on a case-by-case basis since it is a function of both the length and the C/G content of the nucleic acid sequence into primer and template complementary regions.        c) Elongation (60° C.-85° C. for 60-180 seconds). It is the only step actually producing new nucleic add molecules. The temperature at which elongation is performed is a function of the chosen polymerase and, apart from few exceptions, it does not correspond to the hybridization temperature. Usually the elongation of the newly synthetized strand is completed by the commonly used polymerases in one or few minutes, depending on the length of the nucleic acid to be amplified. New priming events cannot take place until such extended strand is not released, at least partially, as a single stranded molecule from the double stranded molecule resulting from the elongation.        
Commonly, in order to obtain a quantity of nucleic acid sufficient for further analysis, these steps are repeated for 15-40 cycles, after which the PCR amplification usually reaches a plateau due to the exhaustion of the polymerase.
Even if many different technologies, strategies, and reaction conditions have been developed on the basic scheme, PCR has severe limitations due to a series of specific requirements:                a) an heating and cooling device;        b) a polymerase which remains highly progressive and accurate even after many cycles at the high temperatures required for the nucleic acid denaturation;        c) the fractionation of the amplification process into many cycles, so that the polymerization is not continuous but synchronized with the thermocycling process. The nucleic acid polymerization reaction cannot be achieved in a continuous way, since it is regularly interrupted to reestablish the conditions necessary for the hybridization between the free primer and template molecules, slowing down the entire process.        
Various solutions have been disclosed in the prior art to overcome such limitations of classical PCR, in particular in providing alternative approaches to allow the strand separation and hybridization in isothermal conditions, without thermocycling.
The Strand Diplacement Amplification (SDA) is an isothermal nucleic acid amplification and detection method which makes use of a polymerase in conjunction with an endonuclease that will cut only the polymerized strand such that the polymerase will displace such strand while generating a newly polymerized strand (EP 497272; Walker G T, PCR Methods Appl 1993, 3 (1):1-6). This technique, based on the repetition of the single strand nicking, extending and displacing steps, has been variously adapted (WO 96/23904; Westin L et al., Nat Biotechnol 2000, 18(2):199-204), but such approach has a limited applicability. An endonuclease must be added together with the polymerase, and therefore the range of allowed temperatures at which the whole process can be performed is restricted to the one maintaining the activity of both enzymes. In the case of endonucleases, such range (usually 25° C.-50° C.) is usually too low to avoid non-specific hybridization of the primers to the template, leading to a considerable proportion of reactions which are nonproductive or generate undesired products. Moreover, the nucleic acids acting as primers and/or templates may have to be additionally modified, since it is mandatory that the endonucleolytic recognition site has to be absent from the template region to be amplified, while it has always to be present into the primer sequence.
The Rolling Circle Amplification (RCA) is a technique making use of a DNA polymerase elongating circularized oligonucleotide probes under isothermal conditions with either linear or geometric kinetics, and generating tandemly linked copies of the DNA molecule to amplify as a consequence of a complex pattern of strand displacement events (Walter N G and Strunk G, Proc Natl Acad Sci USA 1994, 91 (17): 7937-41; WO 94/03624). Also in this case, even if the technology has been variously adapted (WO 97/19193, Isaksson A and Landegren U, Curr Opin Biotechnol 1999, 10(1):11-5), such techniques have several limitations like, for instance, the use of circularized molecules and the production of the amplified molecules not as single entities but as a series of molecules containing a variable number of copies of the original sequences, tandemly linked to each other.
Other techniques have been developed to solve the problem through an intermediate DNA-RNA hybrid in an isothermal, multienzyme reaction containing an RNA polymerase, a ribonuclease and a reverse transcriptase (WO 91/04340, WO 92/08800, WO 97/04126; Gebinoga M and Oehlenschlager F, Eur J Biochem 1996, 235 (1-2): 256-61). It is evident that several problems can be encountered using these methods, related both to the rapid RNA degradation and to the complexity of finding the conditions to make all the components work correctly in the reaction at the same time, maintaining the specificity and the speed of the amplification process.
Even if the amplification technologies are instrumental in the analysis of very small amount of nucleic acid sequences, the throughput of nucleic acid sequencing methods is linked not only to the availability of isolated nucleic acids in an amount sufficient to be processed and obtain reliable information, but also to the possibility to process several samples quickly and with limited human intervention. Most of the genetic information is still now generated using technologies, like gel electrophoresis, which are cumbersome, labour intensive, difficult to automate, and require relatively large devices. Moreover, such methods generally allow only the individual processing of each nucleic acid entity to be sequenced. Therefore, many technologies were developed to provide specific solutions to allow the parallel analysis of several different nucleic acids.
In the prior art, a common way to increase throughput is the processing of many samples in parallel using a solid support, often defined as “DNA microarray” or “DNA chip”, on which nucleic acid are immobilized and then analyzed using different approaches involving either labeled single nucleotides or labeled nucleic acids (Southern E et al., Nat Genet. 1999, 21(1 Suppl): 5-9; Lockhart D J and Winzeler E A, Nature 2000, 405 (6788): 827-36). Large molecules (e.g. molecules over 500 nucleotides long) as well as smaller molecules such as oligonucleotide primers, can be efficiently linked to a solid support in a covalent manner by physical or chemical means, either non-specifically or using a specific chemical group at one end (Adessi C et al., Nucleic Acids Res 2000, 28(20):E87; Okamoto T et al., Nat Biotechnol 2000, 18(4): 43841). The combination of suitable robotics, micromechanics-based systems, and microscopical techniques makes technically feasible the ordered deposition and analysis of up to millions of nucleic acids per cm2 of support.
Using similar approaches, a primer extension reaction can be obtained by immobilizing either the primers or the template molecules (WO 91/13075, WO 00/47767). Alternatively, primers can be grafted to a surface and, in conjunction with free primers in solution, allow the amplification and attachment of a PCR product onto the surface (Andreadis J D and Chrisey L A, Nucleic Acids Res 2000, 28(2): e5). Primers can be also immobilized on a matrix and the template molecules be kept in the liquid phase in a range of temperature (58° C.-74° C.) allowing, in a very random and uncontrolled way, an equilibrium between single/double stranded forms, as well as a single base extension (Dubiley S et al., Nucleic Acids Res 1999, 27(18): e19).
The main disadvantage of these technologies is that, for each application, a DNA chip has to be designed and manufactured first, an operation still quite lengthy, complex, and expensive, and therefore it can be afforded only when very large numbers of the chip are required. Moreover, the chips can be often used only for hybridization and/or single base elongation, are not reusable, and for each chip only one sample of nucleic acids can be processed at each time (Cheung V G et al., Nat Genet. 1999, 21(1 Suppl):15-9; Bowtell D D, Nat Genet. 1999, 21(1 Suppl): 25-32).
Other technologies were recently developed at the scope to solve, at least partially, the problems related to the amplification and the sequencing throughput by coupling the two processes more effectively.
WO 96/04404 (Mosaic Technologies Inc.) discloses methods of detection of a target nucleic acid potentially contained in a sample. The method involves the induction of a PCR based amplification of a target nucleic acid only when the target nucleic acid is present in the sample being tested. Specific primers are attached to a solid support, allowing the amplified target nucleic acid sequences also to be attached to such support. The two strand-specific primers are, as for conventional PCR, specifically designed so that they hybridize sequences flanking, or internal, to the target sequence to be amplified and drive the standard thermocycling process.
The first step in this PCR-based amplification process is the hybridization of the target nucleic acid to the first specific primer attached to the support (“primer 1”). A first elongation product, which is complementary to the target nucleic acid, is then formed by extension of the primer 1 sequence. On subjecting the support to the high temperatures necessary for strand dehybridization, the target nucleic acid is released and can then participate in further hybridization reactions with other primer 1 sequences which may be attached to the support. The first attached elongation product may then hybridize with the second specific primer (“primer 2”), attached as well onto the support and a second elongation product complementary to the first one can be formed by extension of the primer 2 sequence and is also attached to the support. Thus, the target nucleic acid and the first and second elongation products are templates molecules capable of participating in a plurality of hybridization and extension processes, limited only by the initial presence of the target nucleic acid and the number of primer 1 and primer 2 sequences initially present. The final result is a number of copies of the target sequence attached to the surface.
The amplification technique disclosed in this document is generally called as “Bridge Amplification” since, after the first cycle when the template is in solution, both template and primer molecules are immobilized on the support through their 5′ end, forming therefore a typical bridge-like structure when they hybridize.
Since this amplification process allows the immobilization only of the target nucleic acid, the monitoring of the support allows a generically qualitative evaluation of the presence or absence of a specific target sequence predefined by the operator when designing the primers. The Bridge Amplification can be used, therefore, as an high throughput sequence analysis method if different sets of first and second primers are arrayed on different regions of the solid support, and if it is not desired a base-by-base analysis of the target sequence. Moreover, such technology can provide a substantial improvement in the serial analysis of different sequences and/or samples only if primers specific for each different target sequence are known and can amplify the target sequence in similar amplification conditions.
Other technologies try to exploit more effectively the mechanism underlying the Bridge Amplification. WO 98/44151 (Glaxo) discloses how, by engineering all the nucleic acid templates to be analyzed with the addition, at both extremities, of linker sequences complementary to the immobilized primers, each template molecule in solution can be randomly arrayed and amplified, irrespectively of their actual sequence. In this way, identical PCR amplification products are immobilized at high density in a discrete area of the solid support, which is called “DNA colony”, due to the similarity to bacterial colonies when they are observed on a plate. Each DNA colony can be visualized and analyzed singularly, for example, by hybridization with labeled reference sequences or by using a primer elongation-based approach. WO 00/18957 (Appl. Res. Syst.) discloses how the efficacy of the Bridge Amplification technology is improved by immobilizing simultaneously both primer and template molecules on the solid support before that amplification actually begins.
A substantial limitation of technologies based on Bridge Amplification is that, whatever specificity of primer sequences or immobilization strategy is chosen, there is no teaching on how to perform the process in isothermal condition (i.e. without the necessity of a thermocycling process) and therefore, they still suffer from the same limitations associated to PCR. Even if it is possible to apply a reduced number of cycles, due to the sensibility of the analysis and visualization systems, there is the additional complication that dehybridization temperature can affect the uniformity of the solid support, as well as the stability of the nucleic acids binding with the surface. However, none of the isothermal amplification systems disclosed in the prior art achieves, at the same time, the density of different template molecules, which can be amplified and analyzed simultaneously, and the self-sufficiency characterizing the Bridge Amplification technology.
It would be therefore advantageous to devise a method where the primer hybridization, strand elongation and strand separation happen in a continuous fashion, at constant temperature, applying the principles of Bridge Amplification. Actually, the more critical point is to obtain the strand separation, since only the temperature or the use of complex and specific techniques have been disclosed in the prior art to solve the problem. WO 97/47767 (Sarnoff Corp.) suggested the use of chemical or electrostatic based denaturation procedures applied to amplification processes. However, such procedures needs highly dedicated materials, like an electric field generating device, or chemicals, like NaOH, poorly compatible with common amplification procedures and requiring specific additional steps (i.e. modifying primers, washing away the chemicals before restarting the elongation).
The Bridge Amplification technology would provide further advantages if it would be possible to trigger simply the strand separation in isothermal conditions, provided that strand separation would be possible only after that elongation has been completed or, if not completed, without disturbing the nucleic acid polymerase completing the elongation.